Trop. plant pathol. (2018) 43:257–262 https://doi.org/10.1007/s40858-017-0200-2

SHORT COMMUNICATION

Phoma destructiva causing blight of tomato plants: a new fungal threat for tomato plantations in Brazil?

Adans A. Colmán1 & Janaina L. Alves1 & Meiriele da Silva1 & Robert W. Barreto1

Received: 17 July 2017 /Accepted: 24 October 2017 /Published online: 20 November 2017 # Sociedade Brasileira de Fitopatologia 2017

Abstract Solanum lycopersicum is among the most impor- Solanum lycopersicum (Solanaceae) is one of the most impor- tant crops in Brazil. This crop is affected by a large range of tant and broadly grown vegetables in the world. Brazil is fungal diseases that are recognized as major limitations for among the major producers of tomato worldwide (Vale et al. tomato production. Recently, plants grown in a greenhouse 2007). A large range of fungal and oomycete diseases affect in Viçosa, Minas Gerais, Brazil, were found to bear severe tomato (Jones et al. 2014; Lopes and Ávila 2005), some of blight symptoms. A pycnidial coelomycete was repeatedly which are universally regarded as major threats to tomato found in association with necrotic tissues. The fungus had its production, such as early blight, caused by Alternaria spp., morphology recognized as equivalent to that of Phoma and and late blight, caused by Phytophthora infestans. Others, related genera. A phylogenetic analysis based on nrDNA such as Phoma rot, are considered as having less relevance. (ITS) and partial β-tubulin (TUB) sequences led to the con- In Brazil, this disease is considered to be of secondary impor- clusion that the fungus involved was Phoma destructiva. tance (Inouie-Nagata et al. 2016). Although better known for Pathogenicity tests showed that, after 5 days, blight symptoms causing post-harvest disease of tomato fruits, Phoma developed on leaves, flowers and stems of plants belonging to (Didymellaceae) occasionally also affects stems and leaves thirteen different tomato varieties tested. This fungal species is in the field (Kimati et al. 2005;Robletal.2014). mostly known for causing post-harvest tomato rot, which is In July of 2015 several plants grown in a greenhouse in only regarded as a secondary disease in Brazil. This is in the campus of the Universidade Federal de Viçosa (Viçosa, disagreement with the observations made in this work. Here, state of Minas Gerais, Brazil), were found to bear severe the disease symptoms caused by the fungus were very severe, leaf blight symptoms (Figs. 1a, b). Lesions also appeared fully justifying the scientific name of the pathogen. Under on stems, which were similar to those seen on leaves favorable environmental conditions, aggressive strains of (Fig. 1c). All diseased plants died because of the disease. P. destructiva, such as the one isolated in this study, may A coelomycete asexual morph was repeatedly found in become significant threats to tomato plantations in Brazil. association with the necrotic tissues. Direct isolation of the fungus in pure culture was performed through aseptic transfer of spores from colonized tissue onto Keywords Solanum lycopersicum . Coelomycetes . PDA medium with a sterile fine-pointed needle. A representa- Didymellaceae . Pycnidial fungi . Phylogeny . Solanaceae tive isolate was deposited in the culture collection of the Universidade Federal de Viçosa, (Accession number COAD 2069). A representative dried specimen of infected tomato tis- sue was deposited in the herbarium of the Universidade Section Editor: Trazilbo J. Paula Jr. Federal de Vicosa (VIC 44080). Sections of selected fragments of infected leaves bearing fruiting bodies were mounted in * Robert W. Barreto [email protected] lactoglycerol for observation under a light microscope (Olympus BX 51) fitted with an Olympus e-volt 330 digital 1 Departamento de Fitopatologia, Universidade Federal de Viçosa, camera. Biometric data were recorded based on at least 30 Minas Gerais, Viçosa, MG 36570-900, Brazil measurements of various structures. 258 Trop. plant pathol. (2018) 43:257–262

Fig. 1 Phoma destructiva. a. Early stages of infection of Phoma destructiva on a tomato leaf. b. Ibid on flowers. c. Necrotic spots on stem and on a young sprout. d. Tomato plant with severe blight 20 days after inoculation. e. Transversal section of a pycnidium showing a layer of conidiogenous cells. f. Conidia with and without septa. Bars:e=8µmandf=20µm

In order to obtain representative fungal DNA, COAD 2069 Purification Kit (Promega) following the protocol described was grown on PDA at 25 °C under a 12-h daily light regime by Pinho et al. (2012). for one week. DNA was extracted from approximately 40 mg The nrDNA (ITS) and partial β-tubulin (TUB) se- of fungal mycelium using the Wizard Genomic DNA quences of COAD 2069 were amplified using the primer Trop. plant pathol. (2018) 43:257–262 259 pairs ITS1/ITS4 (White et al. 1990) and T1/Bt2b, (Glass Table 1 GenBank accession numbers of Phoma spp. DNA sequences and Donaldson 1995;O’Donnell and Cigelnik 1997), fol- used in the phylogenetic analysis lowing the protocols described by Pinho et al. (2012)and Fungal species Strain number GenBank Acc. No. Sung et al. (2007). Amplification of ITS and β-tubulin, produced sequences of approximately 540 and 567 bp, re- ITS TUB spectively. The nucleotide sequences were edited with the Stagonosporopsis actaeae CBS 106.96 GU237733 GU237670 DNA Dragon software (Hepperle 2011) and deposited in CBS 105.97 GU237734 GU237671 GenBank under accession numbers MF 426960 for ITS P. acetosellae CBS 179.97 GU237793 GU237575 and MF448543 for β-tubulin. Additional ITS and β- P. aliena CBS 379.93 GU237851 GU237578 tubulin sequences used in this study were retrieved from CBS877.97 GU237910 GU237579 GenBank (Table 1). P. digitalis CBS 109.179 GU237744 GU237604 ITS and TUB consensus sequences were compared with CBS 229.79 GU237802 GU237605 others deposited in the GenBank database using the P. destructiva var. destructiva CBS 133.93 GU237779 GU237602 MegaBLAST program. The most similar sequences were CBS 378.73 GU237849 GU237601 aligned using MUSCLE (Edgar 2004) and built in MEGA P. destructiva TS24 KR559677 KU507405 v.5 (Tamura et al. 2011). All of the ambiguously aligned regions within the dataset were excluded from the analy- P. destructiva ICMP 14884 KT309887 KT309473 ses. Gaps (insertions/deletions) were treated as missing da- P. destructiva var. CBS 162.78 GU237788 GU237600 diversispora ta. Bayesian inference (BI) analyses employing a Markov P. bulgarica CBS 357.84 GU237837 GU237589 Chain Monte Carlo method were performed with all se- CBS 124.51 GU237768 GU237590 quences, first with each gene separately and then with the P. cr ys ta ll if er a CBS 193.82 GU237797 GU237598 β concatenated sequences (ITS and -tubulin). P. eu py re na CBS 374.91 FJ426999 FJ427110 Before launching the BI, the best nucleotide substitu- CBS 527.66 FJ427000 FJ427111 tion models were determined for each gene with P. herbarum CBS 615.75 GU237874 GU237613 MrMODELTEST 2.3 (Posada and Buckley 2004). Once CBS 502.91 FJ427022 FJ427133 the likelihood scores were calculated, the models were P. matteuciicola CBS 259.92 GU237812 GU237627 selected according to the Akaike Information Criterion P. omnivirens CBS 991.95 FJ427043 FJ427153 (AIC). The SYM + I + G model of evolution was used CBS 654.77 FJ427044 FJ427154 for ITS, whereas GTR + I + G was used for β-tubulin. P. polemonii CBS 109.18 GU237746 GU237648 The phylogenetic analysis of the concatenated sequences P. saxea CBS 419.92 GU237860 GU237655 was performed on the CIPRES web portal (Miller et al. CBS 298.89 GU237824 GU237654 2010) using MrBayes v.3.1.1 (Ronquist and Heulsenbeck P. insulana CBS 252.92 GU237810 GU237618 2003). The other phylogenetic analyses were conducted as described by Pinho et al. (2012). Sequences derived from P. multirostrata CBS 110.79 FJ427030 FJ427140 this study were deposited in GenBank (http://www.ncbi. CBS 274.60 FJ427031 FJ427141 nlm.nih.gov/genbank)(Table1), and the alignments and CBS 368.65 FJ427033 FJ427143 trees in TreeBASE (www.treebase.org) under entry P. pereupyrena CBS 267.92 GU237814 GU237643 number S21657. P. commelinicicola CBS 100.40 GU237712 GU237593 A pathogenicity test was conducted including thirteen P. costarricensis CBS 506.91 GU237876 GU237596 tomato varieties, namely: Forty, Ikram, Paronset, Platinum, CBS 497.91 GU237870 GU237597 Fusion, Santa Clara Siluet, Aguamiel, Caribe, Serato, Predador, Gladiador, Dominador and Vento. All varieties carry the Mi allele that confers resistance to root knot nem- atode (Meloidogyne spp.). COAD 2069 was grown on Phoma destructiva var. destructiva Plowr., Gard. Chron. II PDA medium at 25 °C for 7 days under a 12-h daily light 16: 621. 1881. regime and a conidial suspension (1 × 106 conidia/mL) was Disease: On leaves, stems and flowers, starting as small prepared for inoculation. Plants were sprayed with this black dots (1 to 2 mm diameter) that became circular to some- suspension until runoff and kept for 2 days in a dew cham- what irregular, slightly depressing the plant tissues and ber at 25 ± 3 °C and later transferred to a greenhouse. In forming zonate necrotic areas, where numerous fruiting bod- addition, two plants were sprayed with sterile water and ies were formed. Lesions coalesced with age and led to entire kept under the same conditions to serve as controls. blight of leaves, flowers and stems. 260 Trop. plant pathol. (2018) 43:257–262

Morphology: Internal mycelium indistinct; external myce- Based on morphology, P. destructiva has been separated lium absent; pycnidia immersed globose to irregular, 60– into two varieties P. destructiva var. destructiva and 150 × 70–180 μm, walls of thick textura angularis, ostiolate, P. destructiva var. diversispora differing in conidial size and brown, smooth; conidiogenous cells enteroblastic, septation (De Gruyter et al. 2002). Phoma destructiva var. ampulliform to doliiform, 4–6 × 3.5–8, hyaline, smooth; co- destructiva is aseptate and has been reported as a common nidia oval to ellipsoidal, 4.5–8 × 2.5–3 μm, mainly aseptate tomato and sweet pepper pathogen, causing fruit rot and small but larger 1-septate consistently produced, biguttulate, hya- black spots on leaves and fruits in the tropics. line, smooth. Phoma destructiva var. diversispora is morphologically dis- In culture: On PDA and PCA, fast-growing (5.5–7.5 cm tinct from var. destructiva by producing larger one-septate diameter after 7 days), edge entire, slightly convex, aerial conidia - a feature found in members of the section mycelium cottony centrally, followed by an area of sparse Phyllostictioides. (Aveskamp et al. 2010), and is considered mycelia, centrally pale olivaceous green alternating with dark to be restricted to Europe (De Gruyter et al. 2002). The “ type mouse grey, periphery composed of immersed mycelium, di- variety” of P. destructiva var. destructiva was assigned to sec- urnal zonation present; slightly humid centrally, olivaceous tion Phoma whereas the other Phoma reported on tomato were black reverse; sporulation abundant on PCA. allocated to other sections: Phoma lycopersici to section Typical leaf blight symptoms developed on leaves, flowers Boeremia and Phoma radicina to section Paraphoma (Chen and stems of all ten tomato varieties five days after inocula- et al. 2015; Gruyter et al. 2012). The ambiguity of some sec- tion. The fungus was isolated from diseased tissues of inocu- tion designations has been discussed by Aveskamp et al. lated plants, producing typical P. destructiva colonies, (2010). Phoma destructiva is an example of such an ambigu- confirming the pathogenicity of COAD 2069 to a range of ity because its two varieties are accommodated in two differ- tomato cultivars (Fig. 1). The morphological features of the ent sections due to the presence or absence of septate conidia fungus isolated from diseased tissue were typical of Phoma (Aveskamp et al. 2010). Based on morphology, COAD 2069 destructiva (Table 2) and this identification was strongly sup- fits into P. destructiva var. diversispora,however,itdidnot ported by phylogenetic analysis of concatenated ITS and TUB group with such a variety in the phylogenetic analysis. These gene sequences (Fig. 2). observations suggest that emphasis put on spore size and Seven species of Phoma have been reported causing dis- septation by taxonomists to decide on the placement of fungi eases on tomato in Brazil (Chen et al. 2015; Farr and Rossman of this group in the past may have been misleading. 2016), including Phoma destructiva, which has been recorded Phoma blight has been an important, although sporadic causing fruit rot only once (Batista and Alves 1981). problem in some countries such as the United States and Nevertheless, that record is incomplete and obscure not sup- India (Jones et al. 2014). Recently, Phoma was recorded ported by any morphological description of the fungus nor causing blight on greenhouse-grown tomato in Malaysia deposit of voucher specimens in herbaria or culture collections. (Rashid et al. 2016). Seed transmission of Phoma diseases

Table 2 Morphology of Phoma species recorded on Lycopersicum esculentum worldwide

Fungi Conidia Section Reference

Septation Size (μm) Shape

COAD 2069 1–septate 4.5–8 × 2.5–3Ovaltoellipsoidal Phyllostictoides?Thisstudy Phoma destructiva var. destructiva aseptate 3.5–6×2–2.5 Oblong to ellipsoidal Phoma Aveskamp et al. 2010 Phoma destructiva var. destructiva 1–septate 8.5–11.5 × 2–3.5 Subglobose to ellipsoidal Phyllostictoides Aveskamp et al. 2010 or allantoid Phoma exigua var. exigua 1–2 septate 7–10 × 2.5–3.5 Subglobose, ellipsoidal Boeremia Chen et al. 2015 to oblong or allantoid Phoma labilis not mentioned 3.5–5.5 × 1.5–2Ellipsoidal Allophoma Chen et al. 2015 Phoma lycopersici aseptate 5–8.5 × 2–3.5 Subglobose to ellipsoidal Boeremia Chen et al. 2015 or allantoid Phoma radicina not mentioned 4–6×2–3 Ellipsoidal to subglobose Paraphoma de Gruyter et al. 2012 Phoma terrestris not mentioned 4–6×2–2.5 Ellipsoidal Setophoma de Gruyter et al. 2012 Phoma tropica not mentioned 3–4×1–2Ellipsoidal Allophoma Chen et al. 2015 Trop. plant pathol. (2018) 43:257–262 261

Fig. 2 Phylogenetic study of Phoma destructiva. Tree inferred from Bayesian analysis based on concatenated ITS and TUB sequences. Species from Brazil are shown in bold face. Bayesian posterior probabilities are indicated at the nodes. The tree was rooted with Stagonosporopsis actaeae (isolate CBS 105.96 and CBS 106.96)

is known to be important in paprika and tomato. Colonized and those associated with post-harvest fruit disease should seed and transplant seedlings are thought to be involved in be conducted in order to clarify whether they represent long distance dissemination of the pathogen (Jones et al. distinct taxa. 2014). It is likely that, in the case of the outbreak observed in Viçosa, infected seeds have served as the source of in- Acknowledgements The authors wish to thank Fundação de Amparo à oculum. Therefore, considering the high level of disease Pesquisa do Estado de Minas Gerais - FAPEMIG, Conselho Nacional de severity observed on the various tomato varieties evaluated Desenvolvimento Científico e Tecnológico - CNPq and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES for financial in this study, it is important to investigate what was the support of the work. original source of P. destructiva and whether it can be disseminated by infected seeds. It is possible that the spread of the pathogen onto plantation areas, particularly under greenhouse conditions, may result in serious losses, References posing yet an additional threat for this highly valuable crop in Brazil, for which disease management is already a con- Aveskamp MM, de Gruyter J, Woudenberg JHC, Verkley GJM, Crous PW (2010) Highlights of the Didymellaceae: a polyphasic approach stant challenge. Additionally, a more detailed comparison to characterize Phoma and related pleosporalean genera. Studies in between isolates of P. destructiva associated with blight Mycology 65:1–60 262 Trop. plant pathol. (2018) 43:257–262

Batista MF, Alves MLB (1981) Doenças do tomateiro na região de O’Donnell K, Cigelnik E (1997) Two divergent intragenomic rDNA Manaus. Fitopatologia Brasileira 6:573–574 ITS2 types within a monophyletic lineage of the fungus Chen Q, Jiang JR, Zhang GZ, Cai L, Crous PW (2015) Resolving the Fusarium are non-orthologous. Molecular Phylogenetics and Phoma enigma. Studies in Mycology 82:137–217 Evolution 7:103–116 Edgar RC (2004) MUSCLE: multiple sequence alignment with high ac- Posada D, Buckley TR (2004) Model selection and model averaging in curacy and high throughput. Nucleic Acids Research 32:1792–1797 phylogenetics: advantages of Akaike information criterion and Farr DF, Rossman AY (2016) Fungal databases, systematic mycology Bayesian approaches over likelihood ratio tests. Systematic and microbiology laboratory, ARS, USDA. Available at: http://nt. Biology 53:793–808 ars-grin.gov/fungaldatabases/.Accessed10June2016 Pinho DB, Firmino AL, Pereira OL, Ferreira Junior WG (2012) An effi- Glass NL, Donaldson GC (1995) Development of primer sets cient protocol for DNA extraction from Meliolales and the descrip- designed for use with the PCR to amplify conserved genes tion of Meliola centellae sp. nov. Mycotaxon 122:333–345 from filamentous Ascomycetes. Applied and Environmental Rashid TS, Sijam K, Nasehi A, Kadir J (2016) Occurence of Phoma – Microbiology 61:1323 1330 blight caused by Phoma destructiva on tomato (Solanum Gruyter J, Boerema GH, van der Aa HA (2002) Contributions towards a lycopersicum) in Malaysia. Plant Disease 100:1241–1242 monograph of Phoma (Coelomycetes)—VI-2. Section – Robl D, Thimoteo SS, de Souza GCCF, Beux MR, Dalzoto PR, Pinheiro Phyllostictoides: outline of its taxa. Persoonia 18:1 53 RL, Pimentel IC (2014) Occurrence of Candida orthopsilosis in Gruyter J, Woudenberg JHC, Aveskamp MM, Verkley GJM, Groenewald Brazilian tomato fruits (Lycopersicum esculentum Mill.) Brazilian JZ, Crous PW (2012) Redisposition of Phoma-like anamorphs in Journal of Microbiology 45:105–109 Pleosporales. Studies in Mycology 75:1–36 Ronquist F, Heulsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic Hepperle D (2011) DNA Dragon 1.4.1 - DNA Sequence Contig inference under mixed models. Bioinformatics 19:1572–1574 Assembler Software. Available at: www.dna-dragon.com. Accessed 16 June 2016 Sung GH, Sung JM, Hywel Jones NL, Spatafora JW (2007) A multi-gene Inouie-Nagata AK, Lopes CA, Reis A, Pereira RB, Quezado-Duval JB, phylogeny of Clavicipitaceae (Ascomycota, Fungi): identification of localized incongruence using a combinational bootstrap approach. Lima MF (2016) Doenças do Tomateiro. In: Amorim L, Rezende – JAM, Bergamin Filho A, Camargo LEA (eds) Manual de Molecular Phylogenetics and Evolution 44:1204 1223 Fitopatologia. Vol. 2. Doenças das Plantas Cultivadas. Ceres, São Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) Paulo, pp 697–732 MEGA5: molecular evolutionary genetics analysis using maximum Jones JB, Zitter TA, Momol TM, Miller SA (2014) Compendium of likelihood, evolutionary distance, and maximum parsimony – tomato diseases and pests, 2nd edn. APS Press, St. Paul methods. Molecular Biology and Evolution 28:2731 2739 Kimati H, Amorim L, Bergamin Filho A, Camargo LEA, Rezende JAM Vale FXR, Jesus Junior WC, Rodrigues FA, Costa H, Souza CA (2007) (2005) Manual de Fitopatologia: Doenças das Plantas Cultivadas, Manejo de doenças fúngicas em tomateiro. In: da Silva JHL, Vale vol 2, 2nd edn. Editora Agronômica Ceres Ltda, São Paulo FXR. (Org.) Tomate: tecnologia de produção. Suprema, Visconde Lopes CA, de Ávila AC (2005) Doenças do tomateiro. Embrapa do Rio Branco, pp 159–197 Hortaliças, Brasília White TJ, Bruns T, Lee S, Taylor J (1990) Amplification and direct Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science sequencing of fungal ribosomal RNA genes for phylogenetics. In: Gateway for inference of large phylogenetic trees. In: Proceedings Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds) PCR protocols: a of the Gateway Computing Environments Workshop (GCE). New guide to methods and applications. Academic Press, San Diego, pp Orleans. pp. 1–8 315–322